The broad, long-term objective of the current research is to reduce the prevalence of problem drinking and related harm in at-risk subgroups of emerging adult women. Prior research has established that specific subgroups of youth are at elevated risk for hazardous drinking; this disparity is particularly striking for young lesbian and bisexual women. The proposed research aims to evaluate risk and protective factors for hazardous drinking in young women, focusing on both self-medication and social influences. Although more is known regarding the relative prevalence of drinking behaviors in at-risk women, there is little research examining risk factors or protective factors for drinking that are specific to this population. Furthermore, few studies have examined possible mediators and moderators of drinking in young at-risk women, especially studies utilizing longitudinal or event-level methodologies. These methodologies are critical in order to reduce recall distortion, adequately examine temporal relationships between variables, and understand the development of hazardous drinking over time. The purpose of the present application is test the role of self-medication and coping motives and the role of social influences and identity salience in predicting drinking behavior among young (age 18-25) at-risk women. We will examine both how these behaviors change over time and will also conduct a smaller exploratory study to examine drinking at the event-level women who drink. To accomplish this objective the study will include 900 women recruited through online networking communities and advertisements. Hazardous drinking and proposed mediators and moderators of drinking will be assessed annually for three years. A subgroup of 100 women who drink at least twice per week will complete two weeks of daily measures annually to examine risk factors and drinking behavior at the daily level. Specific aims are: 1) testing a self-medication model of high risk drinking, where stressors, psychological distress, and drinking to cope are examined in relation to hazardous drinking over time; 2) testing a social influences model of high risk drinking, where social influences, social norms, social motives, and identity salience are examined in relation to drinking hazardous drinking over time; 3) To examine event-level within-person relationships between stress, psychological distress, covariation in social contexts, and situation-specific drinking norms in predicting daily drinking behavior.